Bicalutamide and RO4929097 in Treating Patients With Previously Treated Prostate Cancer: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Bicalutamide and RO4929097 in Treating Patients With Previously Treated Prostate Cancer (NCT01200810), a randomized phase 2 study. The primary outcome is Time to PSA Progression.

2. Clinical Question

The study focused on PSA progression after local therapy. Its stated primary endpoint was time to PSA progression, rather than a direct measure of tumor size or survival. The registry describes the trial as a partially randomized Phase II study. In the randomized phase, Arm I received placebo and Arm II received RO4929097. The arm descriptions then state that patients entered a combination phase with bicalutamide and RO4929097.

The clinical question therefore concerns the initial randomized treatment assignment and the subsequent course of PSA progression. It does not represent a comparison of bicalutamide against no bicalutamide throughout the study, because the registry describes bicalutamide in the combination phase for both groups. Nor does the record establish a completed comparison: it states that data were not collected for the primary outcome after early termination due to lack of study drug. Only three patients progressed during the randomization phase. The registry does not provide a comparative time-to-progression estimate, confidence interval, or test result.

3. Trial Design

The registry describes a randomized, double-masked, parallel-group Phase II trial. It lists enrollment of 10 and a terminated status. The study involved men with rising PSA after definitive local therapy for adenocarcinoma of the prostate. The conditions listed also include recurrent prostate cancer and Stage IV prostate cancer.

Arm I received oral placebo during the randomized phase, on the listed treatment days in repeating 21-day courses, for up to 18 courses in the absence of PSA progression. Arm II received oral RO4929097 during the randomized phase, with the same stated course pattern and progression condition. The descriptions then state that all patients entered a combination phase, receiving oral bicalutamide and oral RO4929097. The registry does not describe a continuing placebo comparison during that phase.

The trial started in 2010-08, with primary completion recorded as 2012-08. It was terminated early due to lack of study drug. The primary-outcome description says data were not collected and notes only three progressions during the randomization phase. Thus the record documents a planned randomized comparison but does not provide a completed primary endpoint analysis. The registry does not report the reason for the planned enrollment, details of the masking procedures, or a fuller account of how the early termination affected follow-up and data collection.

The trial includes a list of secondary outcomes covering PSA response and progression during the combination and observation phases, time to PSA nadir, and safety and tolerability assessed using NCI CTCAE Version 4.0. The registry record does not provide analysis results for these outcomes. Consequently, the design description should be kept separate from evidence of an estimated treatment effect: assignment and treatment plans are described, but comparative outcome results are not reported.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: During the induction phase, if subjects do not have a decline in PSA ≥ 50%, they are taken off study. Subjects who have a PSA decline ≥ 50% are randomized to Arm I or Arm II.

Overall Study

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Data were not collected as the protocol was terminated early due to lack of study drug. Only three patients progressed during randomization phase.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

6. Primary Endpoint

The registry lists the primary outcome as Time to PSA Progression, measured over Up to 12 months.

The primary endpoint was time to PSA progression. It represents the interval from a defined starting point to the recorded occurrence of PSA progression. In a time-to-event analysis, the timing of progression matters, rather than only whether progression occurred by a single cutoff. The registry specifies a comparison between the two groups using a log-rank test for a maximum of 54 weeks; the listed timeframe is up to 12 months.

PSA progression is used in this study as a disease-monitoring outcome. The endpoint addresses whether the timing of a PSA-defined event differs between the randomized groups. It is not the same as overall survival, a direct measurement of tumor burden, or a patient-reported outcome. The registry does not give a PSA progression threshold or a detailed operational definition, so those criteria are not reported in the registry record.

The primary endpoint cannot be interpreted as a completed comparison in this trial. The record says that data were not collected because the protocol was terminated early due to lack of study drug, and that only three patients progressed during the randomization phase. It gives no group-specific event times, estimated treatment contrast, confidence interval, or log-rank result. The endpoint explains what the trial intended to measure, not a demonstrated difference between the assigned treatments.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

The statistical approach named for the primary endpoint is the log-rank test. The registry says that time to PSA progression would be compared between the groups using this test, for a maximum of 54 weeks. The registry does not report a completed test result. The method can be explained as the planned analysis, but it should not be presented as an analysis that was successfully carried out.

What does a log-rank test compare?

In general, a log-rank test compares the pattern of event timing between groups over follow-up. For a time-to-event endpoint, the event is the specified progression event. The test uses information about when events occur, rather than reducing follow-up to a simple yes-or-no status at a single time. It is commonly used when participants can have different lengths of observation or when some are not observed to have the event before follow-up ends.

That general explanation does not establish how this trial handled follow-up. The registry does not describe censoring rules, the starting point for measuring time, or the operational definition of PSA progression. Those details are not reported in the registry record. Without them, the exact endpoint construction cannot be reconstructed from the registration details alone.

What is censoring, in general?

In a time-to-event study, a participant may leave observation, or the study may end, before the event is observed. Such follow-up is often treated as censored: the available information says the event had not been observed up to a particular point, rather than claiming that the event never occurred. This is a general statistical concept, not a statement about how this trial treated individual participants. The registry does not report censoring counts or rules.

Early termination makes this issue especially relevant to interpretation, because the planned follow-up and data collection may not have been completed. The stated reason for termination was lack of study drug. The record says primary-outcome data were not collected and reports only three progressions during the randomized phase. It does not say how many participants were followed for each duration, whether follow-up continued after drug availability ended, or how the planned maximum follow-up was affected.

What would an estimate and confidence interval add?

In general, an effect estimate gives the direction and magnitude of a comparison, while a confidence interval conveys its statistical precision. A broad interval indicates that a range of effects remains compatible with the observed data; a narrow interval indicates greater precision. A hypothesis test and a confidence interval answer related but different questions: the test assesses compatibility with a specified null hypothesis, while the interval communicates uncertainty around an estimate. These concepts do not yield a result for this trial, because the registry provides no primary comparative estimate or confidence interval.

The planned log-rank test alone would not describe the size of an effect. A complete report would ordinarily need to identify the estimated contrast and its uncertainty, as well as the event and follow-up information needed to interpret it. The registry record does not report those results. No p-value should be inferred from the fact that the method was named in the protocol description.

Why do event counts matter?

For time-to-event comparisons, the number and timing of events contribute directly to the available information. A small event count generally limits how precisely a group difference can be estimated. This is a general principle, not a calculation of precision for this trial. Although the record says only three patients progressed during the randomization phase, it does not give their group assignments or event times. The count therefore cannot be converted into a treatment comparison or a precision estimate.

Likewise, enrollment is not interchangeable with the number of observed events. The record lists enrollment of 10, but the primary endpoint information is anchored by the stated progressions and the absence of collected primary data. Enrollment alone cannot show that the planned comparison was estimable or that a null result was found. A lack of a reported result is not evidence that the groups were equivalent.

How should planning illustrations be read?

General sample-size calculations depend on assumptions about the target effect, significance level, allocation, and statistical model. The registry record describes planning illustrations based on an observed effect, with a two-sided significance level of 0.05, equal allocation, and a normal approximation; for time-to-event illustrations, the stated calculation concerns a total number of events. Such figures are hypothetical planning quantities. They are not observed power and do not establish what this trial was designed to detect. The registry does not report the trial's own sample-size justification, so the planning assumptions should not be mistaken for its original design rationale.

When the primary contrast itself is not reported, an illustration tied to an observed effect cannot be interpreted as a summary of the actual clinical result. It should be read only under its stated assumptions. The record supplies no primary estimate or interval with which to evaluate the magnitude or uncertainty of a treatment difference.

Finally, a statistical method written into a registry is not the same as a completed statistical analysis. The listed log-rank method describes the intended comparison. The stated early termination, missing primary-outcome data, and absence of analysis results mean there is no reported log-rank finding to interpret.

10. Multiplicity, Missing Data and Analysis Populations

The registry lists secondary outcomes in addition to time to PSA progression. They include complete response by PSA during the combination phase; PSA progression during the combination phase; time to PSA nadir during that phase; time to PSA progression during the combination phase; time to PSA progression during the observation phase; PSA progression during the observation phase; and safety and tolerability assessed using NCI CTCAE Version 4.0. These entries indicate a range of planned outcome questions, but the registry does not report results for them.

When a study evaluates several outcomes or comparisons, the chance of obtaining an apparently notable result by chance can increase if no adjustment or prespecified hierarchy is used. This is a general statistical concern. The registry record does not report a multiplicity strategy, an outcome hierarchy beyond naming the primary endpoint, or whether any adjustment was planned. It is therefore not possible to assess how multiple outcome tests were intended to be interpreted.

The registry records no analysis entries and states that primary-outcome data were not collected because the protocol ended early due to lack of study drug. It does not give participant-level missingness, reasons for missing measurements beyond the stated early termination, or methods for handling incomplete records. Those details are not reported in the registry record.

The primary outcome description also does not define the analysis population. The registry does not report whether the intended analysis was based on randomized assignment, treatment received, or another population definition. Nor does it specify how participants who did not reach the combination or observation phases would be treated in analyses. These omissions limit the ability to reconstruct the planned estimand or evaluate possible selection caused by early termination.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 5% reporting threshold.

Most frequent non-serious adverse event terms

The registry's safety entries list 5 participants at risk in Arm I and 5 in Arm II. Serious adverse events affected 0 participants in each arm. The record lists no serious-event terms. These entries describe the counts recorded in the registry; they do not provide detailed event narratives, timing, severity grades, or a causal assessment.

For other adverse events, the record lists 5 affected participants in Arm I and 4 in Arm II. These counts should be read as arm-specific registry entries, not as a comparison demonstrating a safety difference. The record does not provide the event terms or clarify the full detail behind these counts. The numbers are not a measure of event frequency per exposure time, and they do not show how many events occurred per participant.

Deaths are not reported in the registry record. The serious-adverse-event entries also do not establish that no adverse events occurred; they state that no participants were recorded as affected by a serious adverse event in either arm. With the trial terminated early and enrollment listed as 10, these safety figures are limited descriptions, not a basis for concluding that either treatment is safer or less safe.

12. What the Result Does, and Does Not, Mean

The record documents a randomized Phase II study of placebo or RO4929097 during an initial phase, followed by a combination phase that included bicalutamide and RO4929097. It identifies time to PSA progression as the primary endpoint and names the log-rank test as the planned comparison. It also reports enrollment of 10, early termination due to lack of study drug, and only three progressions during randomization.

What the record does not provide is a comparative primary-outcome result. There is no reported group-specific time to progression, effect estimate, confidence interval, or p-value. Therefore, the registry does not show whether the initial randomized assignments differed in time to PSA progression, or how precise such a difference might have been. Missing results should not be interpreted as evidence of no difference, equivalence, or benefit.

The distinction between the randomized phase and the combination phase is important. The stated groups differ initially by placebo versus RO4929097. The arm descriptions then indicate a shared combination phase with bicalutamide and RO4929097. The record does not provide an analyzed comparison that separates the effects of those phases or establishes what the later-phase outcomes would mean relative to the randomized treatment assignment.

The safety record offers limited counts: no serious adverse events are recorded in either arm, while other adverse events are listed for participants in both arms. These entries do not compensate for the missing efficacy analysis and should not be used to infer a treatment effect or comparative safety profile. The registry also does not report detailed adverse-event terms or death counts.

Overall, this is a record of a planned clinical comparison that ended before its primary outcome could be reported. It can inform readers about the registered question, treatments, endpoint, early termination, and selected safety counts. It cannot establish whether either initial treatment assignment changed PSA progression or provide a reliable estimate of the size or precision of any difference.

13. Why This Trial Matters Statistically

This trial illustrates why a registered design and a reported result are different kinds of evidence. A registry can identify the intended population, treatment sequence, endpoint, and statistical test. Here, those details clarify that the primary question concerned time to PSA progression between placebo and RO4929097 during the randomized phase, with a subsequent combination phase. But a named endpoint and planned test do not supply an estimate when outcome data are not reported.

Early termination has statistical consequences as well as operational ones. The record attributes termination to lack of study drug and states that primary-outcome data were not collected. It also notes only three progressions during the randomized phase. Without group assignments, event times, and follow-up details, neither the direction nor precision of a group contrast can be assessed. This is a useful reminder that enrollment alone does not determine how much information a time-to-event analysis contains; the event information and its timing matter.

The record also shows why phase-specific treatment descriptions matter. The initial groups were assigned placebo or RO4929097, then the descriptions place both groups in a combination phase with bicalutamide and RO4929097. A reader should not treat that design as a continuous randomized contrast between combination therapy and placebo. The absent comparative results further prevent disentangling the implications of the treatment sequence.

Finally, the registered secondary outcomes and safety counts should not be mistaken for a completed body of evidence. Outcomes can be listed without results, and recorded adverse-event counts do not answer the primary efficacy question. The statistical importance of this trial lies in the distinction between what was planned and what the registry reports as analyzed.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The central limitation is that the registry says primary-outcome data were not collected because the protocol was terminated early due to lack of study drug. Only three patients progressed during the randomization phase. The record does not provide group-specific events, event times, follow-up durations, an effect estimate, a confidence interval, or a log-rank result. The planned primary comparison therefore cannot be evaluated from the registry results.

The endpoint is not fully operationalized in the record. It names time to PSA progression but does not report the PSA threshold, the starting point for timing, or detailed event-adjudication procedures. Censoring rules and participant follow-up are also not described. These details are not reported in the registry record.

There is limited information about the analysis plan. The registry names a log-rank test, but does not report the censoring approach, analysis population, treatment of incomplete data, handling of participants who might not enter later phases, or a multiplicity strategy. The registry also does not report a sample-size justification or interim-analysis plan.

Early termination and the small enrollment listed in the record constrain interpretation, but the registry does not provide enough information to quantify the effect of those factors on a possible treatment estimate. Safety reporting is also limited: serious and other adverse-event counts are listed, but event terms and details are not. Deaths are not reported in the registry record. These limitations leave both the primary comparison and the fuller safety profile unresolved.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Bicalutamide and RO4929097 in Treating Patients With Previously Treated Prostate Cancer — NCT01200810.
• Registry record last updated 2017-12-02; status last verified 2017-10.
• Results first posted 2014-10-23.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

This is a terminated, randomized, double-masked, parallel-group Phase II study of men with previously treated prostate cancer and rising PSA after definitive local therapy. The registered initial comparison was placebo versus RO4929097, followed by a combination phase with bicalutamide and RO4929097. Enrollment was 10.

The primary endpoint was time to PSA progression, with a planned log-rank comparison for a maximum of 54 weeks. The registry states that primary-outcome data were not collected because the study ended early due to lack of study drug, and notes only three progressions during randomization. No comparative primary result is reported. Safety entries record 0 serious adverse events in each arm, while other adverse events affected 5 participants in Arm I and 4 in Arm II. Deaths are not reported in the registry record.